Hello all,
what exactly changed with en_2+2 between the March vs. July versions?
We run both models ran under the same JAR (version 0.7) & identical
settings, but the older March model seems to perform better for us -- it
gives fewer incorrect detections. As an example, "Find me someone who knows
about foreign currency trading platforms" returns "find" as a concept in 0.7
-July, but not in 0.7-March.
Is this a known feature/bug/regression/improvement? Is there a CHANGELOG for
these models?
What other fixes & changes went into the July version -- is it a problem if
we just keep the March one?
Thanks a lot for any insights,
Radim
------------------------------------------------------------------------------
Meet PCI DSS 3.0 Compliance Requirements with EventLog Analyzer
Achieve PCI DSS 3.0 Compliant Status with Out-of-the-box PCI DSS Reports
Are you Audit-Ready for PCI DSS 3.0 Compliance? Download White paper
Comply to PCI DSS 3.0 Requirement 10 and 11.5 with EventLog Analyzer
http://pubads.g.doubleclick.net/gampad/clk?id=154622311&iu=/4140/ostg.clktrk
_______________________________________________
Dbp-spotlight-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/dbp-spotlight-users
